Abstract
Purpose :
Stargardt macular dystrophy is caused by mutations in the Abca4 gene, which at 6.7 kB, is too large for adeno-associated viral (AAV) vector gene replacement. It is the most prevalent form of recessively inherited macular dystrophy worldwide and is therefore highly clinically relevant with no current treatments. DNA and RNA base editors enable correction of pathogenic G>A and T>C mutations. ABCA4 has ~1200 known pathogenic mutations, of which ~63% are editable transition mutations. Here, we compared a Cas13-ADAR RNA base editing system to an SaKKH-ABE8e DNA base editing system to introduce a mutation in the start codon of mouse Abca4 in vitro prior to proof-of-principle in vivo testing. Further, an all-in-one SaKKH-ABE8e containing both the base editor and the guide was compared to the original dual-vector system.
Methods :
Fifteen Cas13b-ADAR guides were screened using a dual luciferase reporter assay containing a fragment of 5’ UTR and exon1 of mouse Abca4. The 3 best-performing guides were used to target the mouse fragment plasmid for transcript analysis. Only 5 DNA editing guides were screened due to PAM sites, so a transcript analysis was conducted from the start. These were done at a 1:1 guide to ABE ratio for the dual-vector system, reflecting AAV conditions. Of these, the two best were tested as an all-in-one SaKKH. Transcripts were analyzed in EditR.
Results :
Screens of Cas13b-ADAR guides indicate that all 15 guides show activity relative to the non-targeting control (NTC). Four guides showed significant levels of editing, G3, G7, G10 and G11, which showed between 20-34% Renilla:firefly. This was confirmed by transcript analysis, where 21-25% of cDNA transcripts showed the mutant G. Dual-vector SaKKH-ABE8e showed high editing, with ~35-45% of DNA transcripts showing a mutant G. The all-in-one SAKKH showed minimal editing in in vitro screens and further optimisation is underway. All 3 constructs are being tested in a heterozygous Abca4 +/- mouse model to better understand the potential of each base editing system.
Conclusions :
Previous meta-analyses show 63% of pathogenic mutations in Abca4 are amenable to DNA and RNA base editing. The findings indicate that both RNA and DNA base editing can target an Abca4 sequence in vitro, with further in vivo testing required. Nonetheless, there is exciting therapeutic potential for base editing systems in targeting Abca4.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.